Online ads can be annoying unless they come just at the right moment.
If you're looking for a new dress, and you see one you really like while reading an article on your favorite news site, you'll probably click on it. Do you really care if it's an ad? Probably not.
At the same time, no one likes the feeling of being watched or tracked. Data privacy has become one of the highest goods in our internet-driven world – at least in terms of value. Consumers are increasingly sensitive to tracking.
Targeted ads live in the vacuum between dress and privacy. Taken too far, they can outrage users and burn a brand. Taken not far enough and they don’t work. But in the middle, they can work out really well for both advertisers and users. So well that they increase conversion rates by 2x and revenue per by up to 2.7x, according to a 2009 paper by Beales.
This article seeks to explain how to get it right as advertiser, and what you can do as a user to protect yourself from aggressive tracking and targeting.
What is targeted advertising?
Before going further, let’s solidify what targeted advertising actually is.
Targeted advertising definition
Targeted advertising describes the use of banner, display, or textual ads that are shown based on demographics, behavior, or consumer history. These ads show up on social networks, news sites, and e-commerce platforms.
When taken to an extreme, targeted ads can be extremely creepy and follow you throughout the internet. Target stepped into that trap in 2012, when the company sent coupons to seemingly pregnant women who showed typical behavior of pregnant women. However, in some cases, they didn't tell their husbands or fathers, which led to hairy situations, as you can imagine, followed by outrage and negative PR.
This is called inferential targeting, targeting systems add create profiles based on user behavior and try to predict their next purchase in the hope to show an ad just in time. The women that were targeted by Target (pun intended) did buy products typical pregnant women would buy. However, as you can imagine, they were not happy with the advertisements they received.
The FTC has taken action ever since: a new regulation demands paid influencer posts to be indicated as such. That’s a good start, but it is expected that targeted advertising will be restrained even further in the future. As awareness for privacy and personal data grows, governing agencies are likely to take action.
More advertisers, such as Facebook, show an AdChoices icon that explains users why they’re seeing the ad to foster transparency and trust. After all, trust is very important, according to a Harvard Business Review study. When users see ads on platforms they trust, they’re more likely to buy.
From a business perspective, targeted ads can reduce the noise that comes with advertising. Normal ads might speak to potential buyers, but have a huge spray effect: all the uninterested and irrelevant users see the ad as well. In an online context, that means tons of wasted ad dollars. Targeted ads address this problem because the internet makes it possible to track users through most of their experience. The result is more efficient spend. There is another solution to the problem, but we’ll come back to it later.
How do targeted ads work?
Third-party trackers consider different user signals to put together a user profile, save the information in a cookie, and pass it on to advertising platforms like Facebook to display a custom ad.
There are many signals to draw from in order to put a profile together:
Social networks lean heavily on user data to populate ads across their platform. If you click “like” on an ad, for example, it might show up with a higher probability for your friends and vice versa. If enough people click on sidebar ads, they also appear in the main feed, which are called engagement ads. In the end, social networks and most publishers live off of ad revenue. Their goal is to display the maximal number of relevant ads possible. Even Amazon has a growing share of revenue from ads, reaching $3.5 billion in Q3 of 2019.
Even datalike geo-targeting, check-ins at local businesses like supermarkets, or items in your shopping basket can be used for targeting. But not all signals carry the same quality. Browsing history can be messy and make it hard to identify intent. Not every Wikipedia article you read shows a purchase intent, but when you visit the product page of that red dress five times a week, it’s a strong signal. Advertisers have to be aware of that.
Past purchases and items that are currently in your basket are much better data points for future purchases. That’s why retargeting, showing ads based on previously visited pages, is one of the most effective advertising methods if done right. Google tracks users’ IP address on over 2 million sites that are part of the search engine’s Display network in a product called Remarketing.
Advertisers can show ads on pretty much any platform:
- Social networks: Facebook, Instagram, Snapchat, Pinterest, LinkedIn
- Search engines: Google
- E-commerce platforms: Amazon, eBay, Etsy
- Question/answer platforms: Quora, Reddit, GrowthHackers
- Publishers and news sites
Does targeted advertising violate data privacy?
Targeted advertising in itself doesn’t violate personal information and data privacy, even though it might seem like that from a user perspective. By accepting the terms of service on platforms like Facebook and the like, users allow these sites to display (targeted) ads.
The concern about targeted advertising stems from the problem with information being passed “behind our backs”. Whether humans like to reveal information by themselves or not doesn’t matter, but who tells it does.
So, what can you do as an advertiser? Trust is one of the most important values of a brand. Burning it with aggressive ads can hurt, but there are some traps that can be easily avoided.
1. Find the right balance between valuable and intrusive information
Avoid tracking data around ethnicity, medical conditions, and relationship status and focus on purchasing behavior, instead. Both, Facebook and Google, forbid targeting users based on personal hardships or sexual interests.
2. Protect the data you have at all costs
Trust can be forever destroyed if sensitive data leaks, especially credit cards and credit scores. Recent breaches from Equifax in 2017 affecting 147 million people or Marriott in 2018 affecting 50 million people had strongly negative consequences and destroyed the reputation of the companies.
3. Give customers the option to opt-out
Opt-out doesn’t have to be black and white. Allow users to opt-out of specific data tracking and choose targeting options. Providing the option to tick check marks on what data to track can be one way to give consumers autonomy and still user online advertisements. Being transparent with tracking can foster trust.
4. Explain why
Inform users why you’re tracking certain data and how it improves your offering and their experience. Either use a popup, sidebar, or highlight it in your terms of service. Nowadays, being transparent about tracking can enhance brand value.
5. Avoid inference tactics
As mentioned, inferential tracking needs to be weighed carefully and takes a lot of data to get the target audience right. Advertisers should lean mostly on products and services bought in the past as indicators for audience segments. This approach should be used sparingly and cautiously.
What can you do as user? The internet is not going to get rid of advertising as a business any time soon. Online advertising is a $220 billion industry. But users can take action to shield themselves from aggressive targeting.
6. Read the terms
If you’re concerned, read through the terms of service of the big platforms Google, Amazon, Facebook, & Co. If the legal language is hard to understand, consider tools like terms of service; didn’t read that translate terms from big companies into user-friendly English for the layman.
7. Use ad blockers
Use ad blockers that stop sites and platforms from showing you ads based on your browsing habits.
8. Use the incognito mode of your browser
The alternative to targeted advertising: contextual ads
Contextual ads don’t need a cookie with tracking information;they look at the context the user is in instead. Examples are the topic of an article the ad is displayed on or the keyword used for a search. This is achieved through ad matching systems that can recognize the content and what it’s about.
Google ads have the advantage of being shown in a clear context and driven by specific search keywords. Social networks are in a more difficult position because they need to construct that intent from many noisy signals. That gives the search engine an edge over social networks.
Targeted ads work when used within reason
Targeted ads can be a solution to common advertising problems for both consumers and advertisers, but only if they’re used in a respectful manner to the user.
For more information on advertising types, check out our breakdown of display advertising – only on G2.